J K. NCBI Handout Series BLAST homepage & search pages Last Update December 31, 2014

Transcription

1 BLST Homepage and Selected Search Pages Introducing the BLST homepage and form elements/functions of selected search pages National Center for Biotechnology Information National Library of Medicine National Institutes of Health epartment of Health and Human Services Background BLST [1] is a suite of programs provided by NCBI for aligning query sequences against those present in a selected target database. The NCBI BLST homepage ( provides an access point for these tools to perform sequence alignment on the web. The BLST Homepage The BLST homepage consists of several sections, each provides a specific set of functions: 1. The common header (), present in most BLST-related pages, provides easy access to other content or functions not directly accessible from the homepage. 2. The right hand column (B) provides links to a list of recently completed results, BLST-related news and search tips. 3. Pages with web forms for submitting searches are listed as links in the body of the BLST homepage. These links are organized into three categories, BLST ssembled RefSeq Genomes (C), Basic BLST (), and Specialized BLST (E). 4. The search box () in the BLST ssembled Genomes section takes the name of an organism as input and suggests a list of candidates. Selecting from the suggested list will locate the best genomic sequence dataset for BLST alignment purposes. G I J K The common BLST header The common BLST header provides a convenient way to navigate among different pages to access different contents or functions. The NCBI logo (G) links to the NCBI homepage ( to allow access to non-blst related functions and content. The Home tab (H) links to the BLST homepage. BLST search results are temporarily saved for 36 hours. The Recent Results tab (I) links to a page that keeps track of recently submitted search requests that have not expired. The Request I uniquely assigned to a submitted search provides a one-click access to that result. H The Saved Strategies tab (J) lists a set of search setups saved earlier. It allows the examination of search settings used, quick re-launch of these searches, and download of specific strategies for sharing or re-use in standalone BLST. The Help tab (K) points to page with a list of links to help documents, tutorials, references and useful download directories on the BLST ftp site. My NCBI [2] is a free account from NCBI, which allows users to customize their site preference and manage their works performed on the NCBI site. Login-related links for My NCBI (L) are at the right. BLST searches performed while logged in to a My NCBI account enables access to BLST search results for their full 36-hour life span through the Recent Results tab. Strategies saved will be saved permanently. C E L B NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014 Contact: blast-help@ncbi.nlm.nih.gov

2 Page 2 BLST homepage & search pages The Recent Results page BLST search results are available for 36 hours. The Recent Results tab displays a list of recently submitted search requests that have not expired. The list is session-specific and will be lost if session cookie is cleared upon browser exit. or this reason, it is recommended that BLST searches be done with an active login to a My NCBI account such as the one shown below. The My NCBI login from the header is shown as an insert at the upper right (). Each result is given as a row in the table. The identifier in the Request I (RI) column (B) provides an one-click access to the search result. The program, Title and atabase column (C) combine to provide a summary for a specific search. atabase restriction applied are indicated by more and the popup upon mouseover (). The save, download and the red x (E) allow saving the search strategy, downloading the search strategy, and removing the search from the list. B C E I G The input box () above the table is for retrieving other results using their assigned RIs, such as those shared among colleagues, used as teaching or demonstration examples, or those with issues encountered and reported to NCBI s blast-help group. H Clicking the Go button without an RI loads the format form (G) where the content (H) and format (I) of results can be adjusted. The Saved Strategies page The Saved Strategies tab (shown below) displays a list of search strategies. The first four columns (J) provide a good summary of the search settings for each saved entry. The view link (K) loads the settings in a search page, while the download link (L) saves the settings in an SN.1 formatted file for use with standalone BLST or reloading on the web services using the Choose ile and View button (M). Clicking the red X (N) removes the entry from the list. M J Contact: blast-help@ncbi.nlm.nih.gov NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014 K L N

3 BLST homepage & search pages Page 3 unctions of BLST search pages There are five BLST search pages, each performs a specific type of sequence alignment. These pages are the foundation for the NCBI BLST service and will be described in more detail. Table 1 below summarizes key aspects of pages. These pages access a set of common databases, a summary of the contents for these databases are given in Table 2. Table 1. Key features of the BLST search pages in the Basic BLST category Search page Query & database combina on lignment type Programs & func ons (default program in bold) nucleo de blast nucleo de nucleo de nucleo de nucleo de megablast: for sequence iden fica on, intra species comparison discon guous megablast: for cross species comparison, searching with coding sequences blastn: for searching with shorter queries, cross species comparison blast Protein blastp: general sequence iden fica on and similarity searches ELT BLST [2] : similarity search with higher sensi vity than blastp PSI BLST: itera ve search for posi on specific score matrix (PSSM) construc on or iden fica on of distant rela ves for a family PHI BLST: alignment with input pa ern as anchor/constraint blastx nucleo de (translated) blastx: for iden fying poten al products encoded by a nucleo de query tblastn nucleo de (translated) tblastn: for iden fying database sequences encoding s similar to the query tblastx nucleo de (translated) nucleo de (translated) tblastx: for iden fying nucleo de sequences similar to the query based on their coding poten al Table 2. Contents of the common BLST sequence databases atabase Type Content nr (nt) default Nucleo de ll GenBank + EMBL + BJ + PB sequences, excluding sequences from PT, EST, STS, GSS, WGS, TS and phase 0, 1 or 2 HTGS sequences, par ally non redundant. refseq_rna Nucleo de Curated (NM_, NR_) plus predicted (XM_, XR_) sequences from NCBI Reference Sequence Project. refseq_genomic Nucleo de Genomic sequences from NCBI Reference Sequence Project. chromosome Nucleo de Complete genomes and complete chromosomes from the NCBI Reference Sequence project. Human G+T Nucleo de The genomic sequences plus curated and predicted RNs from the current build of the human genome. Mouse G+T Nucleo de The genomic sequences plus curated and predicted RNs from the current build of the mouse genome. est Nucleo de atabase of GenBank + EMBL + BJ sequences from EST division HTGS Nucleo de Unfinished High Throughput Genomic Sequences; Sequences: phases 0, 1 and 2 wgs Nucleo de ssemblies of Whole Genome Shotgun sequences. pat Nucleo de Nucleo des from the Patent division of GenBank. pdb Nucleo de Nucleo de sequences from the 3 dimensional structure records from Protein ata Bank. alu_repeats Nucleo de Selected lu repeats from REPBSE, suitable for iden fying lu repeats from query sequences. See "lu alert" by Claverie and Makalowski, Nature 371: 752 (1994). TS Nucleo de Transcriptome Shotgun ssemblies, assembled from RN seq SR data 16S microbial Nucleo de 16S Microbial rrn sequences from Targeted Loci Project nr default Protein Non redundant GenBank CS transla ons + RefSeq + PB + SwissProt + PIR + PR, excluding those in PT, TS, and env_nr. refseq_ Protein Protein sequences from NCBI Reference Sequence project. swissprot Protein Last major release of the UniProtKB/SWISS PROT sequence database (no incremental updates). pat Protein Proteins from the Patent division of GenBank. pdb Protein Protein sequences from the 3 dimensional structure records from the Protein ata Bank. env_nr Protein Protein sequences translated from the CS annota on of metagenomic nucleo de sequences. tsa_nr Protein Protein sequences translated from CSs annotated on transcriptome shotgun assemblies. NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014 Contact: blast-help@ncbi.nlm.nih.gov

4 Page 4 BLST homepage & search pages Elements of the Standard Nucleotide BLST search page The nucleotide-blast link loads the Standard Nucleotide BLST search page. The top of the page (below the common BLST header) contains the breadcrumb indicating the page position in the site hierarchy (), the page title, a set of tabs for quack navigation among the five core BLST search pages (B), plus links to set the page back to default and to bookmark a search page with customized settings (C). The default display of the page contains three sections with the functions described below. Enter Query Sequence The main input box () takes nucleotide query sequences in various B formats [1]. or a single query, Query subrange boxes (E) define a segment of the query to be used in the search. Query sequences saved in a plain text file can be uploaded using the Choose ile but- G ton (). The lign two or more sequences checkbox (G) changes the Choose Search Set sections below to Enter Subject Sequence to allow comparison between the query and the those in the subject input box (H). E H C Choose Search Set BLST database can be selected from the standard list using the pull-down menu (I). search can be restricted to a subset of entries in the selected database using the Organism field by typing the name of the species, K strains, or taxonomic group in the textbox and selecting from the suggested L J list (J). Checking the exclusion box to the right excludes sequences from the selected organism from the search. Multiple organisms can be selected by adding extra input box using the + button. Specific types of relatively low value sequences can be excluded using the checkboxes below (K). or certain databases, entering custom queries in the Entrez Query textbox (L) will restrict a search to entries satisfying the specified criteria. or example, entering biomol_mrna[prop] N 500:1000[slen] will restrict a search to mrn entries 500 to 1000 bases long. I Program Selection Three programs (M and M Table 1) with different speed and sensitivity are N available for nucleotide nucleotide sequence alignment. The default megablast is better for O certain tasks, such as identifying the input query and searching with large genomic query; discontiguous megablast works better in finding related sequences from other organisms; while blastn works better for short input queries and identifying short matches, it also works better for cross-species searches than megablast. Clicking the BLST button (N) submits the search to BLST server for processing. Results will be automatically displayed when completed. lgorithm parameters link (O) opens a normally collapsed section that allows access to other parameter settings, including adjustment of number of alignments saved, search stringencies, scoring systems (score matrix) and gap penalties, as well as query filtering. etailed descriptions are given next. Contact: blast-help@ncbi.nlm.nih.gov NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014

5 BLST homepage & search pages Page 5 Elements of the Standard Nucleotide BLST search page (cont.) General Parameters Parameters in this section specify the search sensitivity. The Max target sequences () sets the maximum database matches BLST saves for a query. The checked Short queries checkbox (B) allows BLST to automatically optimize settings for short input queries 50 nucleotides or less. The Expect threshold (C) filters out matches that are less significant with Expect value above the setting. The Word size () set the size of the initial seed match phase, smaller settings are more sensitive. The Max matches in a query range (E) limits the matches saved to a given region of the query (such as from repeats) so matches to other region of the query are not crowded out. The default setting of 0 means no limit. Scoring Parameters Parameters in this section specify the search sensitivity. The Match/Mismatch Scores () specifies H the reward assigned to exact match and penalty I assigned to a mismatch. The Gap Costs (G) field specifies how gaps introduced in the alignment should be penalized. or megablast, the default is linear, no penalty for opening a gap, while extending a gap assumes a linear penalty proportional to the length of the gap. or both parameters, non-default settings can be selected using the pull-down menu. ilters and Masking Parameters in the section specify whether low complexity sequences and organism-specific repeats should be filtered (H) and whether to filter only at the initial match stage (Mask for lookup table only) or during alignment extension as well(i). Lower case letters in the query (provided as a mixed upper and lower case letters in ST, representing custom features) can also be masked. Elements of the Standard Protein BLST search page The -blast link in the Basic BLST links to the Standard Protein BLST search page. The top of this page has the same tab and links found in the Standard Nucleotide BLST search page (pg. 4) that provide the same functions. The default page display contains three sections with the functions described below. Enter Query Sequence Refer to the description for Standard Nucleotide BLST (pg.4) for details. In addition, checking the lign two or more sequences will change the Program Selection section to leave blastp as the only choice. Choose Search Set Most of the components are similar to the Standard Nucleotide BLST page (pg.3). The main difference is that the database pull-down menu contains a smaller list of databases (J). Program Selection our different programs (K and Table 1) are available to satisfy various search needs. The K default blastp is a general purpose alignment program for identifying a sequence or finding others similar to it. PSI-BLST is for finding more distant relatives and for PSSM construction. PHI-BLST does alignment with a L pattern in the query as a constraint. ELT- BLST is a more sensitive search using conserved domain matches the query to build a PSSM for the match evaluation. More complex searches may require adjustment of other search settings listed under the lgorithm parameters link (L). Parameters specific to blast are described next. C E G B J NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014 Contact: blast-help@ncbi.nlm.nih.gov

6 Page 6 BLST homepage & search pages Elements of the Standard Protein BLST search page (cont.) The lgorithm parameters portion of the Standard Protein BLST search page is organized in a similar manner to that for the Standard Nucleotide BLST. General Parameters This section is the same as that in the Standard Nucleotide BLST (pg. 4). Scoring Parameters C Eight score matrices from two families are available (). The default BLOSUM62 matrix is the best general purpose matrix. or short queries, PM30 is often selected. Each matrix has its own set of supported gap penalties under the Gap Costs pull-down menu (B). Scores for E alignment can be adjusted to account for the bias in sequence composition using various approaches as indicated by the Compositional adjustments setting. Other options including no adjustment can be selected using the pull-down menu (C). ilters and Masking Parameters in the section specify whether low complexity and whether to filter only at the seed lookup stage (). Lower case letters in the query (provided as a mixed upper and lower case letters in ST, representing custom features) can also be masked (E). These settings are not needed when compositional adjustments are used. B Items unique to translated search pages The page layout for translated BLST search pages is the same as Standard Protein BLST. However, they do contain a few programspecific parameters. Translated BLST: blastx In the Enter Query Sequence section, a Genetic code field () is present under the Choose ile button specify the codon table used in the translation of the input nucleotide query. Choose a code appropriate for the source of the query sequence. The remaining sections are the same as the Standard Protein BLST page. Translated BLST: tblastn The page layout is the same as the Standard Protein BLST search page. The key difference is that the atabase field lists available nucleotide databases instead. Translated BLST: tblastx The layout differences are the presence of the Genetic code field (), which is also present in the blastx page, and the databases listed under the atabase pull-down menu. In addition, the main input box in the Enter Query Sequence takes a nucleotide query. Other search pages BLST search pages under the BLST ssembled RefSeq Genomes category differ from these under the Basic BLST category only in the databases they access. The link names clearly indicate the source organism(s) for the database sequences against which the query will be searched. The Specialized BLST category contains different types of search pages. Those using the core BLST programs and the same general layout described about are summarized in Table 3. Table 3. unc on of Specialized BLST pages following the standard layout Page name Search trace archives Search sequences that have gene expression profiles (GEO) lign two (or more) sequences using BLST (bl2seq) Search or nucleo de targets in PubChem Biossay, Search SR by experiment Search RefSeqGene Searching against specialized databases unannotated nucleo de capillary sequencing trace reads from various organisms nucleo de sequences with expression informa on direct comparison of two groups of sequences. This is the Basic BLST page with lign two or more sequences checked nucleo de or sequences with associated chemical ac vity assay data from PubChem raw nucleo de reads from the next genera on sequencing technology nucleo de sequences represen ng selected loci from the RefSeqGene project Contact: blast-help@ncbi.nlm.nih.gov NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014

7 BLST homepage & search pages Page 7 Other search pages (cont.) Other search pages in the Specialized BLST category have non-standard page layouts. These services use BLST or other alignment programs in combination with other tools to accomplish specialized tasks. Table 4 provides a functional summary of these pages. Table 4. unc on of Specialized BLST pages not following the standard layout Page link name Make specific primers with Primer BLST [4, 5] Search immunoglobulins and T cell receptor sequences (IgBLST) [6] Screen sequence for vector contamination (vecscreen) ind conserved domains in your sequence (cds) [7] ind sequences with similar conserved domain architecture (cdart) Constraint Based Protein Multiple lignment Tool [8] Needleman Wunsch Global Sequence lignment Tool unc ons esigning primers using the primer3 algorithm and checking their template specificity using BLST Searching immunoglobin or T cell receptor sequences against germline databases for annota on of the input immunoglobulin sequences Screening input nucleo de sequences against a library of known vector and other ar ficial sequences to iden fy contamina ons Searching an sequence against a database of curated domains for func onal analysis. This search is performed for all blast requests. Iden fying conserved domains present in the input sequence followed by finding other sequences containing these iden fied domains mul ple sequence alignment tool from NCBI. COBLT search can also be launched from any BLST search results. NCBI implementa on of the Needleman Wunch global pair wise alignment tool for nucleo de or queries Cluster multiple sequences together with their database neighbors (MOLE BLST) This tool takes mul ple related input sequences, finds their nearest neighbors from the database using BLST, and then cluster these sequences according to their sequence similari es. Other ways to access NCBI web BLST services In addition to access through a web browser, BLST web services described above, with the exception for those listed in Table 4, can also be accessed using alternative venues. These venues include the -remote option in different standalone BLST+ programs, the RESTful BLST service (known as QBlast or BLST URLPI), and BLST SOP services. The features of these venues are summarized in Table 5 below. Table 5. eatures of available methods to access NCBI web BLST services Venue Web browser Standalone BLST+ ( remote option) [9] RESTful BLST (QBlast, BLST URLPI) [10] BLST SOP services [11] eatures Intui ve: graphical user interfaces and result presenta on Convenience: ease of searching with single or small batch of query sequences Speed: fast turnaround from the distributed compu ng system Versa lity: available op on enables searching against custom sequences Job Limita on: Not meant for high throughput searches with 1 hour CPU me limit ata par on: ccess to different database requires different search pages Comprehensive: more op ons available than on the Web for customizing and fine tuning the search Batch processing: search with large query sequences by submi ng then in smaller batches automa cally Less manual interven on: op on for saving output in various formats Workflow incorpora on: input and output can be integrated in custom workflow Extra requirements: installing standalone BLST+ package and configuring it properly Comprehensive: more available op ons to customize and fine tune the search than the Web Batch processing: search with large query sequences possible through batching Workflow incorpora on: input and output can be integrated in custom workflow Extra requirements: efficient usage requires scrip ng/programming for reques ng URL construc on and result checking Comprehensive: more available op ons to customize and fine tune the search than the Web Batch processing: search with large query sequences possible through batching Workflow incorpora on: input and output can be integrated in custom workflow Extra requirements: efficient usage requires scrip ng/programming for reques ng URL construc on and result checking evia on: XML based BLST Seqlign SN.1 specs deviates from the one generated directly by BLST NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014 Contact: blast-help@ncbi.nlm.nih.gov

8 Page 8 BLST homepage & search pages Technical assistance NCBI provides technical assistance to the BLST user community through its blast-help group. Problem and bug reports, suggestions and feature requests, as well as other questions related to BLST usage should be addressed to the group (blast-help@ncbi.nlm.nih.gov). Submitting detailed information along with the problem report will help expedite the investigation. Information needed when reporting problems encountered during web BLST searches are: description on the goal of the search The RI of the search The detailed error message The search page and settings used along with a summary of the input query, particularly if the RI was not issued CPU related errors are caused when searches exceed the processing time limit. Repeat the search using Edit and resubmit link to get back to the search page with the following adjustments will help resolve the issue: Reduce the number or size of the input query sequence(s), use subsequence for large single query if possible dd database limit using Organism or Entrez query box to search a focused subset Increase the search stringency by using lower Expect value larger Word size ilters and repeat masking lower number for Maximum target sequences or errors occurred from using -remote option of the standalone BLST+ package, as well as standalone BLST+ package for local searches, the following pieces of information should be provided: description on the goal of the search The platform and version of the installed BLST+ package The complete error message The complete command line used summary and a small sample of the input query file BLST server returned RIs if available or RESTful BLST and SOP BLST, the following pieces of information should be provided: description on the goal of the search The platform and relevant code used to call the service The complete error message summary and small sample of the input query file References 1. NCBI BLST: a better web interface. NCBI BLST: a better web interface. Johnson M et. al. Nucleic cids Res Jul 1;36(Web Server issue):w My NCBI Help omain enhanced lookup time accelerated BLST. Boratyn GM, et. al. Biol irect pr 17;7:12. doi: / Primer-BLST: a tool to design target-specific primers for polymerase chain reaction. Ye J, et. al. BMC Bioinformatics Jun 18;13:134. doi: / actsheet: Primer-BLST. ftp://ftp.ncbi.nih.gov/pub/factsheets/howto_primerblst.pdf 6. IgBLST: an immunoglobulin variable domain sequence analysis tool. Ye J, et. al. Nucleic cids Res Jul 1;41 (Web Server issue):w C: specific functional annotation with the Conserved omain atabase. Marchler-Bauer, et. al. Nucleic cids Res Jan;37(atabase issue): COBLT: constraint-based alignment tool for multiple sequences. Papadopoulos JS, garwala R. Bioinformatics May 1;23(9): BLST Help manual QBlast s URL PI User Guide SOP-based BLST Web Services Entrez Programing Utilities Help Blastdbinfo: PI access to a database of BLST databases. NCBIINSIGHT blog entry Contact: blast-help@ncbi.nlm.nih.gov NCBI Handout Series BLST homepage & search pages Last Update ecember 31, 2014